Heightened utilization of mammography has resulted in dramatic increases in the occurrence of breast carcinoma in situ, yet the public health significance of these lesions remains controversial. We propose a multi- center population-based case-control study to evaluate risk factors and biomarkers of breast carcinoma in situ. We will examine the influence of established risk factors for invasive breast carcinoma on BCIS as well as newly identified risk factors, such as alcohol consumption, postmenopausal hormone use, and physical activity. We will evaluate risk factors for BCIS and its major subtypes, lobular carcinoma in situ (LCIS), ductal carcinoma in sine (DCIS), and the aggressive DCIS subtype (comedo-type DCIS). Thus, we will distinguish etiologic factors, including potentially modifiable exposures, that may be related to specific subtypes of biologic significance. In addition, we will be able to compare BCIS risk factors with those identified in our current case- control study of invasive breast cancer. Our unique approach will allow an investigation of risk factors related to tumors of apparently incremental malignant significance. This study will utilize our existing and highly successful consortium of population-based, case-control studies of breast cancer. In the proposed study, we will interview over a 36 month period, 1,570 women with breast carcinoma in situ identified from the WI, MA, and NH state cancer registries. For comparison, about 2,200 population-based controls will be randomly selected from drivers' license lists and from Medicare beneficiary files in each state. This approach will permit the evaluation of two control groups: women with a recent mammographic history, and general population controls. Consenting subjects will participate in a telephone interview. The proposed study, which includes sufficient numbers of BCIS subtypes of disparate malignant potential, provides a unique opportunity to evaluate biomarkers in relation to early disease development. Thus, we propose to evaluate the relationship between BCIS (and its subtypes) to p53 mutations, HER-2/neu amplifications, and NAT2 genotype. These biomarkers are related to invasive breast cancer; our design will allow us to evaluate the extent to which these markers are related to LCIS, DCIS, and the comedo-type DCIS. In addition, we will evaluate the relation between these biomarkers and exposures; this approach may elucidate risk factors involved in early disease development. The findings of this study may have important implications for breast cancer prevention and intervention. In addition, this study will establish the foundation for follow-up studies.